Project Summary Protein aggregation underlies many neurodegenerative diseases, and previous studies indicate that mutations to proteins involved in stress-granule formation can increase the likelihood that neurodegenerative diseases will develop. For example, mutations to FUS have been shown to lead to the development of amyotrophic lateral sclerosis (ALS), a disease that results in the specific degeneration of motor neurons. FUS is an RNA-binding protein (RBP) that carries proline-tyrosine nuclear-localization signal (PY-NLS), and when the NLS is compromised by mutation, FUS accumulates in the cytoplasm. Cytoplasmic accumulation of FUS leads to high local concentrations of aggregation-prone protein, promoting protein misfolding. Our lab has shown that the nuclear-import receptor Karyopherin-b2 (Kapb2) can recognize wild-type FUS, prevent its aggregation, and promote its disaggregation both in vitro and in vivo. Similarly, Kapb2 can dissolve aggregates comprised of other PY-NLS-bearing proteins such as hnRNPA1. However, when the NLS of either protein is mutated (e.g. FUSP525L and hnRNPA1A288S), Kapb2 is significantly less effective. I have found that by making targeted mutations to the cargo-binding interface, I can partially recover the ability of Kapb2 to prevent the aggregation of FUSP525L. Additionally, it has been shown that Kapb2 is capable of trafficking proteins that lack a PY-NLS, suggesting that there are multiple factors that contribute to defining Kapb2 cargo. Based on these preliminary data, I hypothesize that (1) making targeted compensatory mutations to Kapb2 will enable Kapb2 to bind and disaggregate disease-linked protein, and that (2) Kapb2 can be evolved to disaggregate disease- linked cargo which lack or harbor a mutated PY-NLS. To test these hypotheses, my aims are to engineer Kapb2 to (1) disaggregate ALS-linked FUS variants and (2) disaggregate ALS-linked hnRNPA1 variants. I will do this by taking both a structure-based rational approach and an unbiased screen to generate a library of mutant Kapb2 proteins to be tested in vitro and in vivo. These studies will demonstrate that a nuclear-import receptor can be engineered to recognize and disaggregate a diverse repertoire of cargos, including cargos that are misfolded in disease. Furthermore, engineering Kapb2 to recognize disease-linked protein introduces a new strategy for disaggregating pathologically misfolded oligomers using a human protein disaggregation system.